Welcome back! In our previous lessons, we covered basic data retrieval operations like GetItem and BatchGetItem. Today, we’re advancing to more complex retrieval methods—Query and Scan. These operations are vital for efficiently managing and querying large datasets within DynamoDB. We'll be using the Books table for our examples, which has a primary key consisting of year (partition key) and title (sort key).

The Query operation in DynamoDB allows you to retrieve multiple items by specifying exact or conditional matches on the primary key attributes. For example, you can retrieve all books published in a specific year or filter further by titles within that year using sort key conditions.

The Scan operation examines every item in a table and returns all data attributes by default. While powerful, Scan can be inefficient and costly in terms of read capacity units, especially for large tables.

In DynamoDB, the Query operation uses KeyConditionExpression to find specific items. This is a string that outlines certain key conditions.

You can write the KeyConditionExpression string manually:

Python
Copy to clipboard
1response = table.query(
2    KeyConditionExpression='year = :year_val',
3    ExpressionAttributeValues={':year_val': 2018}
4)
5print("Results:", response['Items'])

Alternatively, use the Key helper class to build expressions:

Python
Copy to clipboard
1response = table.query(
2    KeyConditionExpression=Key('year').eq(2018) & Key('title').begins_with('The')
3)
4print("Results:", response['Items'])

This class handles various comparison operators like eq, lt, lte, gt, gte, begins_with, and between.

Additional parameters refine the query:

Python
Copy to clipboard
1response = table.query(
2    KeyConditionExpression=Key('year').eq(2015),
3    ProjectionExpression="#yr, title, author",
4    ExpressionAttributeNames={"#yr": "year"},
5    ConsistentRead=True
6)
7print("Results:", response['Items'])

In this example, ProjectionExpression specifies the attributes to retrieve, while ExpressionAttributeNames solves conflicts between attribute names and DynamoDB reserved words.

A simple Scan reads every item in the entire table and returns all attributes for each item:

Python
Copy to clipboard
1response = table.scan()
2print("Results:", response['Items'])

To more efficiently scan and filter out unnecessary items from the results, use a FilterExpression:

Python
Copy to clipboard
1response = table.scan(
2    FilterExpression="contains(title, :title_val) AND year = :year_val",
3    ExpressionAttributeValues={':title_val': 'Guide', ':year_val': 2017}
4)
5print("Results:", response['Items'])

Filter expressions can use logical operators like AND, OR, and NOT to combine conditions and provide more refined results.

You can also use the DynamoDB boto3 helper class Attr to build the filter expression:

Python
Copy to clipboard
1from boto3.dynamodb.conditions import Attr
2
3response = table.scan(
4    FilterExpression=Attr('title').contains('Guide') & Attr('year').eq(2017)
5)
6print("Results:", response['Items'])

This class offers various comparison methods (like eq, ne, lt, lte, gt, gte, begins_with, between, contains) which let you construct complex filter expressions and save on read capacity by filtering out unnecessary items right in the AWS Cloud.

"Query" Limitations:

• Items Retrieved: You can retrieve up to 1 MB of data in a single Query operation. If your query retrieves more than 1 MB, additional queries are needed to retrieve the rest of the results (pagination).
• Index Usage: Query requires that you use an index (either the primary key or a secondary index).
• Performance: Provides quick access to items using the table's primary key or a secondary index.

"Scan" Limitations:

• Items Retrieved: Like Query, Scan can retrieve up to 1 MB of data per operation. For larger datasets, multiple scans are needed (pagination).
• Resource Consumption: Scans can consume significant read capacity by scanning every item in a table or a secondary index, regardless of the number of items returned.
• Efficiency: Generally less efficient than queries because it does not use indexes to filter results.

Both operations support pagination, but detailed coverage of this topic is beyond the scope of this course.

In today's lesson, you've learned advanced techniques for retrieving multiple items from DynamoDB using Query and Scan operations. These methods enable you to efficiently interact with large datasets, providing flexibility depending on your application's needs.

As you move forward, practice these methods through exercises that challenge you to apply them in various scenarios. This will help you develop a deeper understanding of DynamoDB's capabilities and prepare you for more advanced database management tasks. Keep exploring and enhancing your skills!